| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 From the Department of Health Economics, University of Ulm, Ulm, Germany; and the 2 Department of Ophthalmology II, University Eye Hospital Tübingen, Tübingen, Germany.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
METHODS. An empiric cost-effectiveness analysis was conducted as part of a field study of orthoptic screening. Three-year-old children (n = 1180) in 121 German kindergartens were screened by orthoptists. The number of newly diagnosed cases of amblyopia and amblyogenic factors (target conditions) was used as the measure of effectiveness. The direct costs of orthoptic screening were calculated from a third-party-payer perspective based on comprehensive measurement of working hours and material costs.
RESULTS. The average cost of a single orthoptic screening examination was 12.58 Euro. This amount consisted of labor costs (10.99 Euro) and costs of materials and traveling (1.60 Euro). With 9.9 children screened on average per kindergarten, average labor time was 279 minutes per kindergarten, or 28 minutes per child. It consisted of time for organization (46%), traveling (16%), preparing the examination site (10%), and the orthoptic examination itself (28%). The total cost of the screening program in all 121 kindergartens (including ophthalmic examination, if required) was 21,253 Euro. Twenty-three new cases of the target conditions were detected. The cost-effectiveness ratio was 924 Euro per detected case. Sensitivity analysis showed that the prevalence and the specificity of orthoptic screening had substantial influence on the cost-effectiveness ratio.
CONCLUSIONS. The data on the cost-effectiveness of orthoptic screening in kindergarten may be used by such third-party payers as health insurance or public health services when deciding about organizing and financing preschool vision-screening programs.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Because resources for health care are limited, there is a growing interest among decision makers in the cost-effectiveness of health care interventions.10 For a service to be covered by social health insurance or public health service programs it may become a prerequisite that its cost-effectiveness be demonstrated. Yet, little is known about the cost-effectiveness of preschool vision-screening programs.5 11 12 To our knowledge, no empiric economic evaluations have been conducted to date.
Clinical experience holds that treatment of amblyopia and amblyogenic factors, such as strabismus and certain refractive errors, is most effective when started as early as possible during the sensitive period of visual maturation.13 However, because such visual deficits mostly occur in children with healthy-looking eyes, detection requires examination methods for which the patient’s cooperation is needed. In most children, sufficient cooperation for reliable assessment of monocular visual acuity can be expected from age 3 onward.14 Therefore, the fourth year of life is considered best for vision screening.15 16
In Germany, there is no nationwide specific preschool vision-screening program. Few local public health authorities conduct vision assessment, mainly in school children and of their own accord. Vision assessment is performed by general practitioners (GPs) and pediatricians as part of general preventive care examinations. However, the effectiveness of this assessment in detecting amblyopia and amblyogenic factors is considered to be poor,17 18 because GPs and pediatricians lack the necessary experience for ophthalmic testing in this age group.19 Therefore, only a proportion of affected children have eye disorders detected and treated by age 3, especially those with obvious symptoms (e.g., obvious strabismus) or those who consulted an ophthalmologist for other reasons (e.g., for a "red eye").
Preschool vision screening conducted by orthoptists has been discussed as an option to improve the prevention of amblyopia.20 21 Because of easy access to children, it has been proposed that orthoptic screening be conducted in kindergarten.22 Screening in kindergarten avoids fundamental changes in the health care system that are often difficult to bring about. A dental screening and awareness program conducted in children 6 years of age in German kindergartens has been in place for years and could be used as a model for preschool vision screening.
The cost-effectiveness of screening programs strongly depends on the prevalence of the target conditions, which again depends on exactly how they are defined. To date, there is no consensus on how the most frequent visual developmental deficits due to strabismus and refractive errors should be treated, and at what age, to prevent irreversible damage effectively and efficiently, and to avoid an impairment of health-related quality of life. In a recent review, the prevalence of amblyopia and amblyogenic factors (estimated from the yield of screening programs) was found to range between 2.7% and 4.4%.5 Therefore, cost-effectiveness data should be available that take into account varying levels of prevalence.
The purpose of this study was to analyze empirically the cost-effectiveness of an orthoptic preschool vision-screening program performed on 3-year-old children in German kindergartens to detect untreated cases of amblyopia or amblyogenic factors that, without early treatment, are likely to cause a lasting visual deficit. For comparison, the current situation without an orthoptic preschool vision-screening program was used. Because the prevalence of untreated target conditions may differ in other populations, the impact of changes in the prevalence on the cost-effectiveness was analyzed in detail.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Study Population
One hundred twenty-one (121) kindergartens in two counties of Southern Germany participated in the field study. All 3-year-old children attending these kindergartens were eligible for the study. Parents were asked for informed consent in writing. To our best knowledge, only one parent whose child was already in treatment for strabismus refused participation. On the scheduled day of the orthoptic screening examination, 1180 children were present in kindergarten and were enrolled in the study.
Target Conditions
Target conditions were defined as any untreated visual deficits with a corrected monocular visual acuity of less than 0.5 (10/20) in either eye, or with a corrected monocular visual acuity of less than 0.8 (10/12.5) in both eyes and a logarithmic line difference between eyes of more than two lines on ophthalmologic examination. The number of newly detected cases of target conditions was used as the measure of effectiveness. Cases of visual deficits that had already been treated were not considered when measuring effects.
Components of the Evaluated Screening Program
Because the field study was also designed to evaluate the accuracy of screening, more examinations were conducted than would be components of a screening program. This was taken into account in the cost-effectiveness analysis, which was restricted to the costs and effects generated by the screening program.
A flowchart of the screening program of which the cost-effectiveness was evaluated is shown in Figure 1 . All 3-year-old children were to be examined once by an orthoptist in kindergarten. The orthoptic screening examination consisted of cover tests, examination of eye motility and head posture, and uncorrected monocular visual acuity testing with Lea symbols (Precision Vision, Villa Park, IL). The Lea single optotype test was used at 10 ft (3 m) to test visual acuity, because it combined high testability, reduced test time, and showed similar sensitivity for amblyopia as did line tests.23 The uncorrected visual acuity threshold to pass the examination was set at 0.8 (10/12.5) monocular visual acuity in both eyes, or at least 0.5 (10/20) in both eyes and less than two lines’ difference between visual acuity of the right and left eyes (L. Hyvärinen; Lea-Test Ltd., Helsinki, Finland, personal communication, 1998). Children with a positive screening result would be referred to an ophthalmologist for diagnosis, those with a negative screening result would be rated healthy and those with an inconclusive screening result would be rescreened in kindergarten at a later time when cooperation would be likely to have improved. If the rescreening result was positive, inconclusive, or missing (no participation), children would also be referred to an ophthalmologist for diagnosis.
|
Measurement of Costs
Direct costs of the screening program were measured from the perspective of a third-party payer such as the German social health insurance or public health service that may finance such a program. This means that only those costs were included that would be incurred by a third-party payer, assuming no copayment for medical services by the participants. Nonmedical cost incurred by the parents (e.g., for travel to the ophthalmologist) were not considered.
Labor and material costs of orthoptic screening examinations as well as costs of diagnostic ophthalmologic examinations were considered and calculated in Euro at prices current in 2000. In that year, the average Euro-to-U.S. dollar (U.S.$) exchange rate was 0.92 U.S.$ per 1 Euro, and the average purchasing power adjusted conversion rate was 0.99 U.S.$ per 1 Euro (i.e., close to parity).24
Labor Costs.
Working time of orthoptists and administrative personnel was measured comprehensively in phase I and phase II of the field study. Working time was divided into office time for organizing the screening program and time for visiting the kindergartens. The latter was subdivided into travel time, time for preparing the examination site in kindergarten and examination time. Time measurement was conducted by self-administered working-time registration forms. Orthoptists were asked to fill in the respective form for each visit to a kindergarten. Personnel in charge of organizing the screening program were asked to complete the respective forms once a day.
To obtain labor costs, working time was valued at 23.31 Euro per hour (0.39 Euro per minute), which corresponds to the tariff class Vb of the German Federal Employee Tariff (Bundesangestelltentarif; BAT), which applies to salaried orthoptists working in the public health sector and includes all ancillary wage costs.
Material and Travel Costs.
Material costs comprised costs of orthoptic material, phone, stamps, parent information leaflets and consent forms. Travel costs were valued at 0.27 Euro per kilometer in accordance with German tax regulations.
Costs of Ophthalmologic Examination.
Costs of a standard ophthalmologic examination of children referred from preschool vision screening were calculated at 36.40 Euro, based on the German social health insurance’s relative value scale for outpatient physician services (Einheitlicher Bewertungsmasstab für Ärzte; EBM25
), which defines individual physician services and states point volumes for them. An average conversion factor (point value) of 0.041 Euro was used. For those covered by the social health insurance (approximately 90% of the German population), there is no copayment for physician services. Analogous calculation of costs based on the uniform fee schedule of German private health insurance26
(which is much more detailed) resulted in 40.03 Euro. This value is close and well within the range of costs used in sensitivity analyses.
Sensitivity Analysis
To analyze the stability of results, the influence of 11 parameters on the cost-effectiveness ratio (CER) was evaluated in various sensitivity analyses.
Univariate Sensitivity Analysis.
To evaluate the effect of variations in costs of the single orthoptic examination and of the single ophthalmologic examination, univariate sensitivity analyses were performed by varying these costs upward and downward by 25% each when calculating the costs of the screening program.
Multivariate Sensitivity Analysis Based on a Decision Analysis Model.
To evaluate the influence of other parameters, a decision analysis model of the screening program was developed. The model is based on a spreadsheet (Excel; Microsoft, Redmond, WA), which is provided as freeware by the authors.27
It allows calculation of the CER from model parameters for which the values may be varied. Figure 2
shows the spreadsheet form for input and output of data.
|
|
Analysis 2.
Worst-case and best-case scenarios were analyzed by using least-favorable and most-favorable values for all 11 parameters in the decision analysis model, as shown in Table 1
. For this purpose, values of parameters measured empirically in the study were varied according to their 95% confidence intervals (except for cost data for which the values were obtained in the field study were increased and decreased by 25% each).
Variation of the Screening Program
Analogous to the calculation described herein, the CER of an otherwise identical program was calculated, in which only those 1067 children would be screened whose parents indicated no current ophthalmologic treatment of the child on the consent form.
Ethics
The institutional review board approved the study design, which adhered to the tenets of the Declaration of Helsinki.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
|
For a conservative estimate, the cost calculations for the screening program were based on working time measurement from phase I only. However, the reduced working time in phase II was taken into account by the sensitivity analysis, in which the costs of the orthoptic screening examination were varied downward by 25%.
Orthoptists traveled an average of 20 km per kindergarten (round trip). The sum of travel and material costs averaged 15.78 Euro per kindergarten or 1.60 Euro per examined child. The average total costs of orthoptic screening examinations were 124.33 Euro per kindergarten or 12.58 Euro per examination.
Cost-Effectiveness of the Screening Program
Figure 1
shows the number of children who passed through the different steps of the screening program. It also shows the number of nonparticipating and noncompliant children.
A total of 1180 children were screened, of whom 133 (11.3%) had inconclusive results due to insufficient cooperation. Insufficiently cooperative children were significantly younger than cooperative children (mean age, 40.9 vs. 42.9 months, P < 0.0001, t-test), with 64.7% being younger than 42 months (3.5 years). Of the 133 children with inconclusive screening results, 104 participated in rescreening. Therefore, a total of 1284 (1180 + 104) orthoptic examinations were performed at a total cost of 16,156 Euro (1284 x 12.58 Euro). In addition, 140 ophthalmologic examinations were performed in those children with positive, inconclusive, or missing screening results at a total cost of 5097 Euro (140 x 36.40 Euro). The total cost of the screening program was 21,253 Euro (Table 3) .
|
In 12 of the 1180 children, no conclusive result could be obtained in the screening program, because they did not comply with the referral to an ophthalmologist after positive or inconclusive screening.
The cost of the screening program per child for whom a result could be obtained was 18.20 Euro (21,253 Euro/1,168).
Sensitivity Analysis
Influence of Examination Costs.
If, in univariate analysis, the costs per single orthoptic examination were increased (or decreased) by 25% to 15.73 Euro (or 9.44 Euro), the CER would increase (or decrease) by 19% to 1100 Euro (or 749 Euro) per detected case. If the costs per single ophthalmologic examination were increased (or decreased) by 25% to 45.50 Euro (or 27.30 Euro), the CER would increase (or decrease) by 6% to 979 Euro (or 869 Euro) per detected case.
Influence of Prevalence, Sensitivity, and Specificity.
The CER of 924 Euro per detected case found empirically in the study may result from various combinations of prevalence, sensitivity, and specificity. Figure 3
shows the influence of the prevalence of target conditions for various combinations of sensitivity and specificity of orthoptic examination found by multivariate analysis. It can be seen that prevalence has a strong influence on the CER. For example, if sensitivity and specificity were both set at 80%, the CER ranged from 581 Euro (5.0% prevalence) to 2801 Euro (1.0% prevalence) per detected case. Furthermore, it shows that specificity has a stronger impact on the CER than sensitivity. For example, at a prevalence of 2.0%, the CER would be 800 Euro per detected case if sensitivity and specificity were both 100%. If specificity decreased to 80% (with other parameter values staying constant), the CER would increase to 1146 Euro per detected case. If, instead, sensitivity decreased to 80%, the CER would increase less strongly to 984 Euro per detected case.
|
Variation of the Screening Program
If only those 1067 children were screened whose parents indicated no current ophthalmologic treatment of the child on the consent form, 1165 orthoptic examinations (including 98 reexaminations) would have been performed, as well as 105 ophthalmologic examinations. Because of the smaller number of children screened per kindergarten, the costs per single orthoptic examination would increase to 13.54 Euro. Although the total costs of the screening program would decrease by 1,658 Euro to 19,595 Euro, only 21 cases would have been detected, which results in a CER of 933 Euro per case detected (19,595 Euro/21). The additional costs per additional case detected (incremental CER) of the baseline screening program compared with this variation would be at a favorable 829 Euro per additional case (1658 Euro/2).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Causing costs of almost 50 Euro per kindergarten or 5 Euro per child, the labor time required for organizing the screening program was the single most important resource used. The costs of organization were substantially higher than assumed in a cost-effectiveness analysis based on a decision analysis model using data from the literature and estimated data,11 which explains much of the differences in results. It is likely that the time for organizing the screening program could be reduced when the program became routine. In this respect, the costs and CER reported can be considered conservative estimates.
Screening in kindergarten provided an easy access to children, requiring only a little effort by the parents. Thus, from a societal perspective, indirect costs of screening in kindergarten can be expected to compare favorably with other screening options, because parents do not have to take extra time off to have their children screened.
Robustness of the Results
Sensitivity analysis showed that the prevalence of untreated target conditions and the specificity of orthoptic screening had substantial influence on the CER. Because the prevalence of untreated target conditions may differ in other populations and when different definitions of target conditions are used, the CER was modeled for varying levels of prevalence. Interested readers may test the influence of the prevalence as well as other model parameters on the CER by using the spreadsheet provided by the authors.27
The influence of the prevalence was the main reason for the large range of the CER in the worst-case-best-case scenarios.
Program Variation
In our sample it was more cost-effective to screen all children, regardless of the information provided by the parents on current ophthalmologic treatment of the child. This was because the prevalence of untreated target conditions among those children whose parents indicated current ophthalmologic treatment was almost equal to the prevalence among all other children. Two untreated children with a severe visual deficit had been examined by ophthalmologists before the screening, but the ophthalmologists’ recommendations for a follow-up had been neglected by the parents. Thus, the screening may serve to reinforce the need for intervention and increase compliance with follow-up protocols.
Choice of End Point and Methodology
As in many published economic evaluation studies of screening programs and diagnostic tests in other areas of medicine, in this cost-effectiveness analysis only the costs and effects were considered that occurred up to the clinical end point diagnosis. For a more comprehensive evaluation of orthoptic vision screening, further studies should analyze the costs and effectiveness of treatment and the disability caused by visual deficits.5
The definition of visual deficits could be linked to meaningful reductions of health-related quality of life for that purpose. The generic term amblyopia, which summarizes visual deficits of very distinct pathophysiologic origins, could be omitted in this perspective, as in this study. To make different studies comparable, dedicated software modules could be used to standardize results. The spreadsheet software developed for this study could be refined and used as a reference procedure for this purpose.
In this cost-effectiveness analysis, the alternative course of action with which orthoptic screening was compared was no orthoptic screening, which was associated with no costs and no effects until the chosen end point. In current practice, some of the cases detected by the screening program might have been detected by other means (and at other costs) at a later time had there been no orthoptic screening in kindergarten. To include this in an empiric study would require a far more extensive study design covering several years of individual follow-up. However, the chosen study design is very likely to yield a conservative estimate of the cost-effectiveness of orthoptic screening compared with study designs with a longer follow-up. Most other means of vision assessment tend to be less sensitive, less specific, and more costly than orthoptic screening, which would cause incremental CERs of orthoptic screening to be more favorable than the CER found in this study. In a recent evaluation of various methods of vision assessment performed by pediatricians in Germany,18 the most favorable values found for sensitivity, specificity, and the proportion of inconclusive results were 54%, 78%, and 3%, respectively. By feeding these parameter values into the decision-analytic model introduced earlier,27 the following can be shown: If all children in this study received pediatric vision assessment instead of orthoptic screening, the additional costs of orthoptic screening per additional case detected (incremental CER) compared with pediatric vision assessment would be less than the 924 Euro found in this study as soon as pediatric vision assessment cost more than only 4 Euro per child, assuming a prevalence of 2.5% as an example. This is mainly due to the low specificity of pediatric vision assessment causing many costly false-positive results. Thus, extending the empiric analysis to a time frame that includes a longer period of follow-up would be likely to yield an even more favorable incremental CER than found in this study.
In conclusion, this study provided data on the cost-effectiveness and its determinants of orthoptic vision screening in kindergarten. This information may be used by third-party payers such as health insurance or public health services when deciding about organizing and financing preschool vision-screening programs. Although we did not analyze the cost-effectiveness of treatment in this study, decision makers should take into account that treatment after early detection through screening potentially avoids a visual deficit that may last for almost a lifetime.
|
Footnotes |
|---|
Submitted for publication January 29, 2002; revised May 13, 2002; accepted May 29, 2002.
Commercial relationships policy: N.
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked "advertisement" in accordance with 18 U.S.C. 
1734 solely to indicate this fact.
Corresponding author: Hans-Helmut König, University of Ulm, Department of Health Economics, Helmholtzstrasse 22, D-89081 Ulm, Germany; hans-helmut.koenig{at}mathematik.uni-ulm.de.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
jcbarry/goe.html. Accessed May 17, 2002
This article has been cited by other articles:
|
|
 |
|
  H.-H. Konig and J.-C. Barry Cost-Utility Analysis of Orthoptic Screening in Kindergarten: A Markov Model Based on Data From Germany Pediatrics, February 1, 2004; 113(2): e95 - 108. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C Williams, K Northstone, R A Harrad, J M Sparrow, and I Harvey Amblyopia treatment outcomes after preschool screening v school entry screening: observational data from a prospective cohort study Br. J. Ophthalmol., August 1, 2003; 87(8): 988 - 993. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J-C Barry and H-H Konig Test characteristics of orthoptic screening examination in 3 year old kindergarten children Br. J. Ophthalmol., July 1, 2003; 87(7): 909 - 916. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
